hypoepiluminescence microscopy of pigmented skin lesions ... · dermoscopy hypoepiluminescence...
TRANSCRIPT
DERMOSCOPY
Hypoepiluminescence Microscopy of Pigmented SkinLesions: New Approach to Improve Recognition ofDermoscopic Structures
DOMENICO PICCOLO, MD,� MARIA CONCETTA FARGNOLI, MD,�
GERARDO FERRARA, MD,y GIAN PIERO LOZZI, MD,� DAVIDE ALTAMURA, MD,�
TERENZIO VENTURA, MD,z SERGIO CHIMENTI, MD,y AND KETTY PERIS, MD�
BACKGROUND Hypoepiluminescence microscopy (HELM) is a new dermoscopic approachfor analysis of pigmented skin lesions (PSLs) after surgical excision.
OBJECTIVES The objective was to verify whether this method could provide additionalmorphologic information for diagnostic or didactic purposes compared to conventionalepiluminescence microscopy (ELM).
PATIENTS AND METHODS Thirty-one PSLs from 30 patients were consecutively evaluatedby ELM and, after excision, by HELM. For HELM examination, the lesion was positioned ona glass slide and illuminated from above with a halogen lamp and from underneath withan LED source. All lesions were subsequently examined histopathologically.
RESULTS In 11 of 31 (35.5%) lesions, a typical pigment network, as assessed by ELM,appeared bidimensional with HELM. In 9 lesions (9/31; 29%) ELM showed a gray-blue area,while HELM allowed us to distinguish 5 lesions (5/9, 55.5%) with gray area predominantshowing a lichenoid lymphocytic infiltration and few melanophages from the other 4 le-sions (4/9, 44,5%) with heavy dermal accumulation of pigmented melanocytes or mela-nophages where a blue area was clearly visible at HELM. In 9 other cases (29%), ELManalysis revealed a central homogeneous dark brown/black pigmentation that in 7 caseswas seen under HELM examination to consist of globules.
CONCLUSIONS HELM is particularly useful in evaluating heavily PSLs or structureslocated in the reticular dermis.
The authors have indicated no significant interest with commercial supporters.
In the past two decades dermo-
scopy [dermatoscopy, epilumin-
escence microscopy (ELM), skin
surface microscopy] has grown
enormously from a research tool
to a standard technique for in vivo
diagnosis of pigmented skin lesions
(PSLs).1–15 Therefore, an import-
ant limitation of dermoscopy is
that this technique cannot be used
to investigate structures in the mid
and deep dermis. For example,
dermoscopy differentiates poorly
between the blue veil of regression
structures, histopathologically re-
lated to melanophages in the upper
dermis, and the pigmentation due
to the presence of typical or atyp-
ical melanocytes in the dermis. The
recognition of the blue structures
might be related with dermo-
scopist’s experience. Furthermore,
two recent studies16,17 demon-
strated that the presence of
& 2006 by the American Society for Dermatologic Surgery, Inc. � Published by Blackwell Publishing �ISSN: 1076-0512 � Dermatol Surg 2006;32:1391–1397 � DOI: 10.1111/j.1524-4725.2006.32311.x
1 3 9 1
�Department of Dermatology, University of L’Aquila, L’Aquila; yDepartment of Pathology, GeneralHospital of Benevento, Benevento; zDepartment of Pathology, General Hospital of L’Aquila, L’Aquila;yDepartment of Dermatology, University of Tor Vergata, Rome, Italy
dermoscopic signs of regression
may predict histopathologic
disagreement.
In 2000, Braun and colleagues18
introduced a new technique for the
study of PSLs known as hypolu-
minescence microscopy (HLM).
This technique involves visualizing
the morphologic structures of a
surgically excised PSL using illu-
mination from the dermal side.
In our study we combined classic
ELM with HLM to create hypo-
epiluminescence microscopy
(HELM). The aim was to deter-
mine whether this method could
provide additional morphologic
information for diagnostic or
didactic purposes, compared to
conventional ELM.
Materials and Methods
Thirty-one PSLs observed at the
outpatient clinic of the Depart-
ment of Dermatology, University
of L’Aquila, from January 1996 to
May 2005 were included in the
study. Selection criteria included:
(1) equivocal clinical findings
(based on ABCDE criteria) and/or
(2) equivocal dermoscopic fea-
tures (irregular pigment network,
gray-blue pigmentation, irregular
dots, and globules and streaks).
Every PSL was consecutively
evaluated by ELM and, after
excision, by HELM.
After covering the lesion with
immersion oil, we performed
classic ELM in vivo using a digitalFigure 1. Stereotypical ELM (left) pigment network that appeared bidimensionalwith HELM (right).
Figure 2. (A) ELM analysis (left) revealed a central homogeneous dark brown/black pigmentation that consists of globules at HELM examination (right). (B)Large and deeply pigmented nests of melanocytes within the junction and/orbeneath the epidermis (hematoxylin and eosin; original magnification, � 100).
D E R M AT O L O G I C S U R G E RY1 3 9 2
H E L M F O R E VA L U AT I N G P S L
camera (Coolpix 990, Nikon
Corporation, Tokyo, Japan) with a
special dermoscopy objective con-
taining a halogen lamp (Nevu-
screen, Arke s.a.s., Avezzano,
Italy). The lesion was visualized on
a 15-in. high-resolution monitor,
and the image was captured in
JPEG format under high-quality
compression (2048�1536 reso-
lution) and stored in our computer
database (Nevuscreen). The mag-
nification of the images varied
from 22 to 66� . The excised le-
sion was placed in a drop of alco-
hol on a glass slide, with the digital
camera of the ELM system located
above the slide. The lesion was il-
luminated from the dermal side
using a dermatoscope (Dermlite,
3Gen LLC, Dana Point, CA) with
an LED source. We used the same
ELM system for visualization and
storage of the HELM images, re-
producing similar magnification
and orientation of ELM images to
permit their comparison.
Results
Thirty-one PSLs were excised
from 18 women and 12 men,
ranging in age from 18 to 76 years
(mean, 39 years). Eighteen of 31
PSLs (58.1%) were located on the
trunk, 5 (16.2%) on the lower
extremities, 3 (9.7%) on the but-
tocks, 2 (6.4%) on the upper ex-
tremities, 2 (6.4%) on the acral
sites, and 1 (3.2%) on the face.
Histopathologic examination
identified 25 melanocytic nevi (23
Clark neviF14 dysplastic, 5
compound, and 4 junctional nevi
Fand 2 Spitz/Reed nevi), 3 mel-
anomas, 1 lentigo simplex, 1 irri-
tated seborrhoeic keratosis, and 1
actinic keratosis.
HELM revealed additional mor-
phologic features that were not
apparent with ELM analysis, par-
ticularly regarding pigment net-
work, gray-blue and dark brown/
black pigmentation. Eleven of 31
lesions (35.5%) showed a typical
ELM pigment network that ap-
peared bidimensional with HELM
(Figure 1). ELM analysis of 9
other lesions (29%) revealed a
central homogeneous dark brown/
black pigmentation that in 7 cases
was seen under HELM examin-
ation to consist of globules (Figure
2A) corresponding histopatho-
logically to large and deeply pig-
mented nests of melanocytes
within the dermal-epidermal
junction (Figure 2B) and/or be-
neath the epidermis. In the re-
maining 2 cases (32.3%), HELM
showed pigment network and
black dots (Figure 3A) that cor-
responded to pigmentation of the
basal layer of the epidermis and
melanin in the stratum corneum
(Figure 3B).
Figure 3. (A) Presence of peripheral black lamella at ELM examination (left) thatcorresponded to network and black dots with HELM (right). (B) Pigmentation ofthe basal layer of the epidermis and melanin in the stratum corneum (hema-toxylin and eosin; original magnification, � 100).
3 2 : 1 1 : N O V E M B E R 2 0 0 6 1 3 9 3
P I C C O L O E T A L
Visualization with ELM showed a
gray-blue pigmentation in nine
other cases (29%) evaluated with
both ELM and HELM. In five of
these nine cases (55.5%) HELM
examination showed a gray area
(Figure 4A) corresponding histo-
pathologically to a heavy lichen-
oid lymphocytic infiltration
together with few melanophages
in the upper dermis, suggestive of
regression in an early inflamma-
tory phase (Figure 4B). In the
other four cases (44.5%), HELM
disclosed a blue-whitish area
(Figure 5A) that corresponded
histopathologically to the
presence of melanophages or
pigmented melanocytes in the
dermis (Figure 5B).
In the remaining two cases (6.4%)
ELM showed only a globular
pattern (Figure 6A), whereas
HELM revealed the presence of a
pigmented network combined
with globules (Figure 6B). Histo-
pathologic examination support-
ed the HELM analysis, showing
pigmentation of the epidermal
basal layer and nests of melano-
cytes in the superficial dermis
(Figure 6C).
Discussion
At present, dermoscopy is con-
sidered fundamental for the study
of PSLs. Some clinically equivocal
benign PSLs (particularly atypical
Spitz and Clark nevi) that can be
difficult to distinguish clinically
from melanoma, however, remain
difficult to differentiate from
melanoma also dermoscopically.
These problems might be due to
the intrinsic difficulties of evalu-
ating these types of lesions, which
have dermoscopic features similar
to those of melanoma, as well as
to technical limitations.
Heavily pigmented or nodular
lesions often represent the ‘‘gray
zone’’ of dermoscopy. In this
study we used a new technique,
named HELM, to evaluate ‘‘gray-
zone’’ PSLs after surgical excision.
Two different sources of illumin-
ation (a halogen lamp above and
an LED source underneath the
excision specimen) were specific-
ally chosen to better differentiate
between epidermal and dermal
dermoscopic structures. Images
obtained with this technique
showed better contrast than im-
ages obtained using a single type
of illumination (i.e., halogen-
halogen, LED-LED), and the der-
moscopic features became more
visible.
Braun and coworkers18 demon-
strated earlier that HLM involves
Figure 4. (A) Gray-blue pigmentation with ELM (left) that appeared as a grayarea with HELM (right). (B) Heavy lichenoid lymphocytic infiltration togetherwith few melanophages in the upper dermis, suggestive of regression in anearly inflammatory phase (hematoxylin and eosin; original magnification,�100).
D E R M AT O L O G I C S U R G E RY1 3 9 4
H E L M F O R E VA L U AT I N G P S L
visualizing the morphologic
structures, such as globules, of a
surgically excised PSL using illu-
mination from the dermal side to
make the dermal structures more
visible comparing with ELM. In
our study, in 9 of 31 cases (29%),
HELM analysis showed that
black-brown pigmentation that
appeared homogenous by ELM
was actually composed of glob-
ules (7 cases) or network and
black dots (2 cases). In our esti-
mation, the combination of both
types of illumination probably
creates a special filter that over-
comes some of the limitations
arising from the use of classic
ELM illumination or HLM alone.
The most confounding dermosco-
pic pattern is the so-called gray-
blue area, also known as blue-
whitish veil, blue veil, gray-blue
pigmentation, peppering, and
regression structures. Zalaudek
and associates17 demonstrated
that this pattern reflects multiple
aspects of regression. Invasive
melanoma can also have the same
dermoscopic features, however,
particularly the blue-whitish veil,
with blue areas arising not from
regression but from the presence
of atypical melanocytes in the mid
and deep dermis. Argenziano and
colleagues19 showed that agree-
ment among experts using
classic pattern analysis was only
fair for the regression structures
(k = 0.44) and poor for the
blue-whitish veil (k = 0.32).
Although dermoscopically similar,
these structures need to be
distinguished for proper
management of the lesions.
In this study we observed a gray-
blue pigmentation in 9 of 31
(29%) cases with ELM, but only
5 of these 9 cases (55.5%) showed
a gray area with HELM. Histo-
pathologic examination of these
5 cases revealed a lichenoid
infiltrate with few melanophages
in the upper dermis (suggestive of
regression in an early inflamma-
tory phase). In the remaining
4 cases (44.5%), a blue-whitish
area was clearly visible with
HELM and was histopathologic-
ally related to the presence of
sheets of deeply pigmented
melanocytes and/or melanophages
in the dermis. Therefore, from a
practical point of view, HELM
allows better differentiation of
‘‘blue-gray areas’’ from ‘‘blue-
whitish veil,’’ with the former
mainly composed of lymphocytes
(or, possibly, representing fibrosis)
and the latter mainly composed of
sheets of pigmented melanocytes
and/or melanophages. This find-
ing helps to predict if a suspicious
melanoma according to ELM
could be a thick (blue-whitish
veil) or a thin (blue-gray areas)
melanoma using HELM.
These promising results, although
preliminary and based on a
small number of samples, should
inspire further HELM studies,
Figure 5. (A) The gray-blue pigmentation of the ELM images (left) was visualizedas blue-whitish at HELM examination (right). (B) Presence of melanophages orpigmented melanocytes in the dermis (hematoxylin and eosin; original magni-fication, � 100).
3 2 : 1 1 : N O V E M B E R 2 0 0 6 1 3 9 5
P I C C O L O E T A L
above all on clinically or
dermoscopically equivocal PSLs,
to evaluate the utility of this
method in distinguishing the blue
veil of regression structures from
that of deep melanophages–
melanocytes in the dermis.
In summary, HELM provides ad-
ditional information compared to
ELM allowing to distinguish the
dermoscopic structures located at
different layers of the epidermis.
A diffuse pigmentation at ELM
can demonstrate the presence of a
pigment network and/or dots and
globules using HELM. Moreover,
HELM permits us to differentiate
the bluish pigmentation of re-
gression from that of melano-
phages or sheets of melanocytes in
the dermis.
The appearance of dermoscopic
structures under HELM is also
important for didactic purposes.
Visualization of some PSLs using
double illumination brought out a
pigment network that was not
evident with classic ELM and
further showed a bidimensional
aspect to the network, which ap-
peared to be raised. In addition,
the shape, size, and colors (black
or brown) of the black dots and
the brown globules were clearly
visible, and they appeared to be
located in two different vertical
plans, with the black dots more
superficial than the brown glob-
ules. These observations were
confirmed by the histopathologic
correlates.
Using HELM we might better re-
late dermoscopic features to their
histopathologic correlates. For
example, differentiating between
black dots and brown globules or
blue structures can often be diffi-
cult or impossible in routine
practice, but in many cases this
problem can be solved with
HELM. On the basis of our re-
sults, we can recommend HELM
as a useful bridge between classic
ELM and histopathology.
Acknowledgment We are very
grateful to Barbara J. Rutledge,
PhD, for critical review and edit-
ing assistance.
References
1. Pehamberger H, Steiner A, Wolff K. In
vivo epiluminescence microscopy of pig-
mented skin lesions. I. Pattern analysis of
pigmented skin lesions. J Am Acad Der-
matol 1987;17:571–83.
2. Soyer HP, Smolle J, Hodl S, et al. Surface
microscopy: a new approach to the
diagnosis of cutaneous pigmented tu-
mors. Am J Dermatopathol 1989;11:
1–11.
3. Bahmer FA, Fritsch P, Kreusch J, et al.
Terminology in surface microscopy. J Am
Acad Dermatol 1990;23:1159–62.
4. Kenet RO, Kang S, Barney JK, et al.
Clinical diagnosis of pigmented lesions
using digital epiluminescence micro-
scopy: grading protocol and atlas. Arch
Dermatol 1993;129:157–74.
Figure 6. (A) ELM image showed a globular pattern. (B) HELM examinationrevealed the presence of a pigmented network combined with globules.(C) Pigmentation of the epidermal basal layer and nests of melanocytes in thesuperficial dermis (hematoxylin and eosin; original magnification, � 100).
D E R M AT O L O G I C S U R G E RY1 3 9 6
H E L M F O R E VA L U AT I N G P S L
5. Pehamberger H, Binder M, Steiner A,
Wolff K. In vivo epiluminescence micro-
scopy. Improvement of early diagnosis of
melanoma. J Invest Dermatol
1993;100:356–62.
6. Soyer HP, Smolle J, Leitinger G, et al.
Diagnostic reliability of dermoscopic
criteria for detecting malignant mela-
noma. Dermatology 1994;190:25–30.
7. Menzies SW, Ingvar C, Crotty KA, Mc-
Carthy WH. Frequency and morphologic
characteristics of invasive melanomas
lacking specific surface microscopic
features. Arch Dermatol 1996;132:
1178–82.
8. Argenziano G, Soyer HP, De Giorgi V,
et al. Interactive atlas of dermoscopy. In:
Dermoscopy: a tutorial (book) and CD-
ROM. Milan: Edra Medical Publishing
& New Media, 2000.
9. Argenziano G, Soyer HP. Dermoscopy of
pigmented skin lesionsFa valuable tool
for early diagnosis of melanoma. Lancet
Oncol 2001;2:443–9.
10. Soyer HP, Argenziano G, Talamini R,
Chimenti S. Is dermoscopy useful for the
diagnosis of melanoma? Arch Dermatol
2001;137:1361–3.
11. Kittler H, Pehamberger H, Wolff K,
Binder M. Diagnostic accuracy of der-
moscopy. Lancet Oncol 2002;3:159–65.
12. Binder M, Schwarz M, Winkler A, et al.
Epiluminescence microscopy. A useful
tool for the diagnosis of pigmented skin
lesions for formally trained dermatolo-
gists. Arch Dermatol 1995;131:286–91.
13. Yadav S, Vossaert KA, Kopf AW, et al.
Histopathologic correlates of structures
seen on dermoscopy (epiluminescence
microscopy). Am J Dermatopathol
1993;15:297–305.
14. Soyer HP, Kenet RO, Wolf IH, et al.
Clinicopathological correlation of pig-
mented skin lesions using dermoscopy.
Eur J Dermatol 2000;10:22–8.
15. Massi D, De Giorgi V, Carli P, Santucci
M. Diagnostic significance of the blue
hue in dermoscopy of melanocytic le-
sions: a dermoscopic-pathologic study.
Am J Dermatopathol 2001;23:463–9.
16. Ferrara G, Argenziano G, Soyer HP, et al.
Dermoscopic and histopathologic diag-
nosis of equivocal melanocytic skin le-
sions: an interdisciplinary study on 107
cases. Cancer 2002;5:1094–100.
17. Zalaudek I, Argenziano G, Ferrara G,
et al. Clinically equivocal melanocytic
skin lesions with features of regression: a
dermoscopic-pathological study.
Br J Dermatol 2004;150:64–71.
18. Braun RP, Saurat JH, Krischer J. Hypo-
luminescence microscopy of pigmented
skin lesions. Melanoma Res
2000;10:141–4.
19. Argenziano G, Soyer HP, Chimenti S,
et al. Dermoscopy of pigmented skin
lesionsFresults of a consensus meeting
via the Internet. J Am Acad Dermatol
2003;48:679–93.
Address correspondence and reprintrequests to: Ketty Peris, MD,Department of Dermatology, Univer-sity of L’Aquila, Via Vetoio-Coppito2, 67100 L’Aquila, Italy, or e-mail:[email protected].
3 2 : 1 1 : N O V E M B E R 2 0 0 6 1 3 9 7
P I C C O L O E T A L